Season analytics
What the model actually did, graded after the fact. Pregame snapshots are never rewritten.
Graded calls
—
Season to date
Hit rate
—
Recommended sides only
Units
—
Flat 1u staking
ROI
—
Return on amount staked
MAE
—
Mean absolute projection error
Avg CLV
—
Line value vs close
Weekly performance
Hit rate and units by week.
No weekly rows yet
The dashed line is 52.4% — roughly break-even at standard -110 pricing.
Projection error
Mean absolute error by week. Lower is better.
No error data
Confidence buckets
High confidence should outperform low. If it doesn't, confidence is broken.
| Bucket | n | Hit | Units | ROI |
|---|
Prop type performance
Where the model earns and where it should stay out.
| Market | n | Hit | MAE | ROI |
|---|
Calibration
A 60% call should win about 60% of the time.
No calibration data
Signals working / not working
Average edge captured per signal, graded after the fact.
| Signal | n | Hit | Avg edge | Verdict |
|---|
Biggest misses
Largest absolute projection errors. Worth reading before trusting a similar call.
Loading graded projections…
These results come from the labelled demo dataset. They demonstrate the grading pipeline, not a real betting record.